**** BEGIN LOGGING AT Wed Nov 28 02:59:59 2012
Nov 28 08:58:25 <dwmw2_gone> how well do we support multiple build targets from the same source tree?
Nov 28 08:58:41 <dwmw2_gone> I can't build in parallel, but if I just switch configs back and forth is that going to work relatively well?
Nov 28 08:58:50 <dwmw2_gone> or can we do a proper out-of-source-tree build?
Nov 28 09:06:21 <russell--> dwmw2_gone: it works reasonably well, except if you are building different subtargets (e.g. x86) with different compiler options, i got bit by the difference between alix2 and soekris net45xx recently
Nov 28 09:07:40 <dwmw2_gone> thanks. Just swapping configs around, you mean?
Nov 28 09:07:48 <russell--> yes
Nov 28 09:08:33 <russell--> i do brcm47xx, atheros, ar7, x86 all in the same tree
Nov 28 09:09:37 <Kaloz> dwmw2_gone: and if you don't mix platforms in a single arch
Nov 28 09:10:34 <Kaloz> dwmw2_gone: and softfloat vs hardfloat :)
Nov 28 09:28:43 <dwmw2_gone> that's oK. I'm doing x86 geos and ar71 wndr3800
Nov 28 10:41:37 <dwmw2_gone> after adding patches, what is supposed to make sure the kernel source tree is rebuilt?
Nov 28 10:42:40 <luka> make target/linux/{clean,compile}
Nov 28 10:51:15 <dwmw2_gone> ta
Nov 28 10:53:53 <dwmw2_gone> Suppose if I want fixes in AA I'd best backport them to 3.3.8...
Nov 28 10:54:16 <[florian]> yes
Nov 28 10:54:28 <[florian]> you are thinking about your atm/br2684 patches?
Nov 28 10:54:34 <dwmw2_gone> yeah
Nov 28 10:54:40 <dwmw2_gone> just building now
Nov 28 10:54:56 <dwmw2_gone> I can no longer panic my 3.6.7 kernel just by killing br2684ctl while it's busy transmitting :)
Nov 28 10:55:09 <[florian]> cool
Nov 28 10:55:17 <dwmw2_gone> and its Ethernet doesn't crap itself either, when I flood it
Nov 28 10:55:27 <dwmw2_gone> well, it does. But at least it recovers after the tx timeout :)
Nov 28 10:55:38 <[florian]> yeah I also saw these patches, thanks for doing this
Nov 28 10:55:47 <dwmw2_gone> I ought to update wholesale to AA. I need to make netifd work first though.
Nov 28 10:55:58 <dwmw2_gone> is there an easy way to switch between netifd and the old setup?
Nov 28 10:56:06 <dwmw2_gone> or do I need to find a second CF card for this?
Nov 28 10:56:19 <[florian]> it's not exactly simple no
Nov 28 10:56:42 <dwmw2_gone> There's "simple" and there's "easier than physically swapping out the storage on the box that currently connects me to the outside world"
Nov 28 10:57:13 <[florian]> swapping rootfs is the easiest way right now
Nov 28 10:59:22 <dwmw2_gone> hm, don't think I can even fit two rootfses on a single card
Nov 28 10:59:35 <dwmw2_gone> perhaps that's the way forward though — put them both on one and hot-swap between them
Nov 28 11:02:58 <dwmw2_gone> hm, or can it cope with rootfs on usb storage?
Nov 28 11:13:16 <[florian]> it certainly can do this, just check your kernel options
Nov 28 11:29:41 <dwmw2_gone> hm, I do have about 8M free on this CF card
Nov 28 11:29:56 <dwmw2_gone> a clean image might just fit. Definitely if I can enable compression. Perhaps btrfs will work...
Nov 28 11:33:32 <dwmw2_gone> very odd
Nov 28 11:33:36 <dwmw2_gone> /dev/loop0         48377  8482     37438  19% /mnt/spare
Nov 28 11:33:39 <dwmw2_gone> [root@westmere spare]# rmdir lost+found/
Nov 28 11:33:43 <dwmw2_gone> /dev/loop0         48377  6014     39906  14% /mnt/spare
Nov 28 11:33:59 <dwmw2_gone> the lost+found directory in a freshly-created image was huge
Nov 28 11:34:05 <dwmw2_gone> and empty, at the same time :)
Nov 28 12:07:26 <dwmw2_gone> hm, that almost worked
Nov 28 12:07:37 <dwmw2_gone> netifd brought up the pppoa, and teql still works through scripts
Nov 28 12:07:40 <dwmw2_gone> or something like that
Nov 28 12:07:54 <dwmw2_gone> except I was returned to the 20th century
Nov 28 12:07:58 <dwmw2_gone> only Legacy IP and no IPv6
Nov 28 13:53:47 <OpenWrt-Commits> jow r34399 packages/utils/dosfstools/Makefile * dosfstool: change download mirror (#12285)
Nov 28 17:58:42 <dwmw2_gone> aha, genext2fs makes a lost+found directory with a size of 5% of the image.
Nov 28 17:58:51 * dwmw2_gone adds -m0 to the genext2fs invocation
Nov 28 18:51:21 <rektide> [florian]: any suggested work arounds for https://dev.openwrt.org/ticket/12539#ticket ?
Nov 28 18:51:33 <rektide> the undefined reference to `lzma_properties_decode@XZ_5.0' error?
Nov 28 18:51:56 <rektide> i've been waiting around for a month hoping my build would magically heal itself
Nov 28 18:52:08 <rektide> (it usually does: the people in this channel rock!)
Nov 28 19:02:13 <jow_laptop> rektide: can you run "ldd /usr/lib/libxml2.so" and paste the output?
Nov 28 19:04:22 <jow_laptop> rektide: also try this patch... http://pastebin.com/JwVP7gMj
Nov 28 20:18:52 <rektide> jow_laptop: liblzma.so.5 => /lib/x86_64-linux-gnu/liblzma.so.5 (0x00007eff5e007000 is the lzma entry from that ldd. gist is: https://gist.github.com/4164002
Nov 28 20:29:51 <rektide> will reply with the status of that patch. make is happening.
Nov 28 20:57:25 <rektide> i believe this is working. there's a lot of stuff compiling. :)
Nov 28 20:58:43 <Chocks> \o/
Nov 28 21:03:45 <OpenWrt-Commits> jow r34400 trunk/package/libs/gettext-full/Makefile
Nov 28 21:03:45 <OpenWrt-Commits> gettext-full: use shipped libxml for the host build, fixes lzma references from creeping in on certain distributions
Nov 29 00:29:44 <SmugLeaf> [  5%] Building C object CMakeFiles/netifd.dir/main.c.o
Nov 29 00:29:44 <SmugLeaf> In file included from /home/weedy/projects/openwrt/build_dir/target-mips_r2_uClibc-0.9.33.2/netifd-2012-10-29/main.c:23:0:
Nov 29 00:29:47 <SmugLeaf> /home/weedy/projects/openwrt/build_dir/target-mips_r2_uClibc-0.9.33.2/netifd-2012-10-29/netifd.h:24:29: fatal error: libubox/ustream.h: No such file or directory
Nov 29 00:29:50 <SmugLeaf> compilation terminated.
Nov 29 00:29:52 <SmugLeaf> make[6]: *** [CMakeFiles/netifd.dir/main.c.o] Error 1
Nov 29 00:29:54 <SmugLeaf> ok, i've been stuck for 3 days now
Nov 29 00:29:58 <SmugLeaf> what gives?
Nov 29 01:05:12 <OpenWrt-Commits> luka r34401 trunk/package/boot/uboot-kirkwood/Makefile * [boot] uboot-kirkwood: build in $(BUILD_DIR)
**** ENDING LOGGING AT Thu Nov 29 02:59:59 2012